Rules of Origin in the World Trading System

1
Rules of Origin in the World Trading System

•  
• Antoni Estevadeordal
• Kati Suominen
• Integration, Trade and Hemispheric Issues
  Division
• Integration and Regional Programs Department
• Inter-American Development Bank
• PREPARED FOR THE SEMINAR ON
• REGIONAL TRADE AGREEMENTS THE WTO
• WORLD TRADE ORGANIZATION
• NOVEMBER 2003

2
Facilitating or hindering trade the role of
rules of origin in RTAs

• I. Rules of Origin Key Issues
• II. (a) Product-Specific Rules of Origin
• II. (b) Regime-Wide Rules of Origin
• III. (a) Measuring Restrictiveness of
  Product-Specific RoO
• III. (b) Constructing a Facilitation Index of
  Regime-Wide RoO
• IV. Empirical Analysis to Evaluate the Impact of
  RoO
• V. Policy Conclusions Is Harmonization the Key
  to Mitigating RoOs Negative Effects?

3
Rules of Origin Key Issues
4
Objectives of RoO

• Two types of RoO non-preferential and
  preferential
• The justification for preferential RoO is to
  ensure that non-members do not obtain access to
  regional preferences (avoid trade deflection)
• However, RoO can be a powerful trade policy
  instrument
• RoOs effects in the S/R different than in the
  L/R
• RoO can fully insulate an industry from the
  consequences of an FTA
• RoO can protect intermediate good producers by
  favoring intra-PTA supply links
• RoO can be used to attract investment in
  strategic sectors
• Very limited theoretical and empirical work
• Details matter a lot!

5
Rules of Origin and WTO

• Non-Preferential Rules of Origin are being
  harmonized under Uruguay Round Agreements
• Preferential Rules of Origin would be a key
  element of discussion under Doha mandates on
  Regional Agreements (Rules Negotiations)
• We also agree to negotiations aimed at
  clarifying and improving disciplines and
  procedures under the existing WTO provisions
  applying to regional trade agreements. The
  negotiations shall take into account the
  developmental aspects of regional trade
  agreements.

6
Product-Specific RoO
7
Types of Product-Specific RoO

• Wholly obtained or produced
• Where only one country enters into
  consideration in attributing origin
• Substantial transformation
• Where two or more countries have taken part in
  the production process
• Change in Tariff Classification (CTC)
• Requires the product to change its HS tariff
  heading, chapter, heading, sub-heading, or item
  in the originating country
• Import Content Rule (MC) or Regional Value
  Content (RVC)
• Requires a MIN of local value added in the
  originating country (or a MAX of value
  originating in non-member countries)
• Technical Requirement (TECH)
• Prescribes that the product must undergo
  specific manufacturing processing operations in
  the originating country

8
Frequency of Various Product-Specific RoO
Criteria
Source World Trade Organization (2002).
9
Product-specific RoO in Europe

• The product-specific RoO of the EUs PANEURO
  system cover more than 65 PTAs, both across
  Europe and in EUs recent extra-regional PTAs
  with Mexico, Chile, and South Africa
• PANEURO model mainly combines CTC (usually CH)
  with RVC and/or TECH

10
Product Specific RoO in the Americas

• Four RoO families can be identified
• 1. LAIA, Andean Community, Caricom
• -across-the-board CH or VC
• 2. NAFTA, Mexicos FTAs, Can-Chi, US-Chi,
  Chi-Cacm
• -high degree of variation in RoO across
  products
• -different types of CTC in combinations with
  VC
• and/or TECH
• 3. Mercosur and its FTAs with Chile and
  Bolivia
• -mainly CH also CH and VC or TECH
• 4. CACM
• -falls between NAFTA and Mercosur models
  mainly CTC but also combinations

11
Product-specific RoO in Asia and RoW

• RoO are based on an across-the-board VC
  criterion
• (usually RVC of 25-50 percent)
• Alternative, across-the-board RoO
  usually involves a change in heading or change
  in subheading
• More complex RoO regimes emerging
• - SADC
• - Japan-Singapore FTA
• - US-Singapore FTA
• - Chile-Korea FTA
• - Some moves to renegotiate RoO in Africa

12
Product-specific NON-PREFERENTIAL RoO

• Harmonization work carried out since
  July 1995 by the WTO and the World Customs
  Cooperation Council
• The most common RoO is change in
  heading however, also important amounts of
  sectoral selectivity
• Some 90 (mainly political rather than
  techncical) issues affecting about 20 of tariff
  lines have yet to be resolved
• Main sticking points
• RoO in machinery CTC- or RVC-based?
• Disagreements over related trade policy issues
  (SPS, EEZ, trademarks, geographic indications,
  etc.)
• Disagreements over role of RoO in anti-dumping

13
Distribution of CTC Criteria by Agreement
14
Type and Height of RVC CriteriaSelected PTAs
15
RoO Combi-nations Selected PTAs
16
Regime-Wide RoO
17
Types of Regime-Wide RoO

• Provisions adding leniency to RoO
• De minimis
• Roll-up or absorption principle
• Cumulation
• Self-certification
• Provisions that may make RoO more restrictive
• Lists of operations insufficient to confer origin
• No-drawback rule
• Complex certification methods
• Inefficient or non-transparent verification by
  customs

18
Frequency of General RoO Provisions
Source World Trade Organization (2002).
19
Regime-Wide RoOSelected PTAs
20
Certifi-cation MethodSelected PTAs
21
Measuring the Restrictiveness of
Product-Specific RoO
22
Measuring and Comparing Restrictiveness of RoO
Regimes

• Restrictiveness measured by Estevadeordals
  (2000) index ranging from 1 (least restrictive)
  to 7 (most restrictive)
• Construction of index
• - Degree of CTC CC more restrictive than CH, CH
  more restrictive than CS, etc.
• - Exceptions to CTC, RVC and TECH add to
  restrictiveness

23
Restrictiveness of RoOSelected PTAs
24
Sectoral Restrictiveness of RoO by HS Sections,
Selected FTAs
25
Profiles of Sectoral Restrictiveness of EU,
NAFTA, and SADC RoO by HS Sections
26
Comparing the Restrictiveness of RoO Regimes

• Sectoral restrictiveness is very similar across
  the main regimes
• Non-preferential RoO echo the PANEURO and NAFTA
  models
• Highest restrictiveness values in sectors with
  relatively high tariffs especially in EU and the
  US
• Agriculture
• Textiles and apparel
• Transportation equipment
• Trade-weighted restrictiveness of RoO deviates
  little from unweighted restrictiveness

27
Constructing a Facilitation Indexof
Regime-Wide RoO Provisions
28
Regime-Wide RoO A Facilitation Index

• Comprised of mechanisms that add flexibility to
  the application of the product-specific RoO
• Based on a scheme that considers five components
  
• De minimis
• Diagonal cumulation
• Full cumulation
• Drawback
• Self-certification

29
Facilitation Index for Selected PTAs
30
RoO Innovations

• Further, it is important to consider innovative
  ad hoc features in some regimes that can
  alleviate the impact of stringent RoO
• - differential application of RoO due to
  differences in the partners
  development levels (esp. drawback)
• - phase-in of the level of value content RoO
  (esp. footwear, vehicles)
• - permanent sectoral deviations from standard
  RoO (esp. from the PANEURO model in EU-Mexico
  and EU-Chile FTAs)
• - TPLs where RoO revoked on a certain amount of
  imports (esp. NAFTA-based regimes)
• - flexibility in the calculation of RVC (esp.
  when partner lacks domestic processing, as
  in Singapores FTAs)
• Stage1             Stage 2
                   Stage 3
• Singapore ? Foreign Country ? Singapore ?
  Exported
• Conventional RoO ? Stage 3 Local Content
• Recognition of OP ? Stage 1 Stage 3
  Local Content

31
Measuring the Impact of RoO
32
Summary of Empirical Evidence

• Important growing pool of theoretical evidence
  restrictive RoO distort trade, investment, and
  production patterns
• Increasing number of empirical studies assessing
  the economic impacts of preferential RoO on
• Firms administrative costs
• Degree of utilization of RTAs (or GSP) conferred
  trade preferences
• Trade patterns in final and intermediate goods

33
Summary of Empirical Evidence RoO and Trade
deflection

• RoO are used legitimately as instruments to
  curb trade deflection
• Estevadeordal (2000) Sectoral restrictiveness of
  NAFTA RoO positively and significantly related to
  MFN tariff differentials between US and Mexico
• Sanguinetti (2003) Sectoral restrictiveness of
  MERCOSUR RoO positively and significantly
  related to MFN tariff differentials between
  Argentina and Brazil
• However, RoO are also used for political economy
  purposes
• The restrictiveness of RoO in NAFTA(Estevadeordal
  2000), Mercosur (Sanguinetti 2003), and EU
  (Suominen 2004) appears to be driven by the same
  political economy variables as drive tariffs

34
Summary of Empirical Evidence Firms
Administrative Costs

• Koskinen (1983) Administrative costs of
  certifying origin in
• EC-EFTA FTA 1.4 5.7 of value export
  transactions
• Herin (1986) Administrative costs of
  certifying origin in EFTA
• 3 5 of value export transactions
• Holmes and Shephard (1983) Average
  export transaction EFTA
• to the EC requires 35 documents and
  360 copies
• Today, official certification of RoO
  for a single shipment costs 7 in Chile, 6-20
  in Brazil, but free in general in EU countries
• Cadot et al. (2004) NAFTAs non-RoO
  related administrative
• costs approximate 2 of the value of
  Mexican exports to the US

35
A Summary of Empirical Evidence Utilization
of Trade Preferences

• Estevadeordal and Miller (2002)
  Canadas utilization rates of preferences to US
  market declined in sectors where NAFTA RoO
  became more stringent than CUFTA RoO
• Cadot et al. (2002, 2004) the mere 64
  utilization rate of NAFTA preferences due in
  part to restrictive RoO
• Brenton and Manchin (2003) low
  utilization rates of the EUs trading partners
  in the textile sector owe to the excessive
  stringency of EU RoO
• Inama (2004) Evidence on low
  utilization rates due to restrictive RoO in GSP,
  ACP, and AGOA

36
Rules of Origin and Utilization RatesCanada-US
FTA vs. NAFTA
Percent of total U.S. imports from Canada
37
Summary of Empirical EvidenceRoO and Trade
Patterns

• Cadot, Estevadeordal and Suwa (2003)
  Mexican exports to the US have been undermined
  by restrictive NAFTA RoO
• Appiah (1999) CGE model shows NAFTA RoO
  distort trade flows and undercut welfare
• James (2004) NAFTA preferences and
  restrictive RoO have undercut Asian textile and
  apparel exports to US
• Flatters and Kirk (2004) Restrictive
  SADC RoO work against efficiency gains that
  members would reach through extra-regional
  outsourcing
• Augier, Gasiorek and Lai-Tong (2004)
  with no cumulation, trade has been 50 lower
  than expected the impact is particularly
  notable in intermediate goods

38
Estimating the Effects of RoO on Trade Patterns
A Gravity Model Approach

• Estevadeordal and Suominen (2004) Estimate a
  gravity model for 156 countries, 100 PTAs and 20
  years (1981-2001)
• Preliminary results restrictive product-specific
  RoO
• (1) curb aggregate trade and
• (2) divert trade in intermediates
• Flexible regime-wide RoO facilitate trade, and
  hence counter the effect of restrictive
  product-specific RoO

39
Gravity Model EstimatesRoO and Bilateral
Aggregate Trade
Ln(Bilateral Trade) b0 b1ln(GDP_PRODij)
b2ln(DISTANCEij) b3ln(BORDERij)
b4ln(COMMON_LANGUAGEij)
b5ln(COLONIAL RELATIONSHIPij)
b6ln (COMMON COLONIZERij)
b7ln(FTAij) b8ln(RoO_RESTRICTIV
ENESSij) b9ln(RoO_FACILITATIONij)
e
40
Regression Results Aggregate Trade and
RoO (Cross-Section)
41
Regression Results Aggregate Trade and
RoO (Cross-Section)
42
Regression Results Aggregate Trade and RoO
(Panel 1981-2001)
43
Regression Results Aggregate Trade and RoO
(Panel 1981-2001)
44
RoOs Effects on Trade in Intermediate
Goods
Ln(Bilateral Trade in Inputs) b0
b1ln(GDP_PRODij) b2ln(DISTANCEij)
b3ln(BORDERij) b4ln(COMMON_LANGUAGE
ij) b5ln(COLONIAL RELATIONij)
b6ln (COMMON COLONIZERij)
b7ln(RoO_REST. FINAL Goodsij)
b8ln(RoO_FACILITATIONij) e
45
RoOs Effects on Trade in Intermediate
Goods Key Variables

• Dependent variable total imports at the 6-digit
  level of intermediate products products
  classified as intermediates by section of the
  Harmonized System and that are used particularly
  intensely in the HS section in question (as
  opposed to being used in other sectors)
• Examples man-made filaments in textile products
  conveyor belts in vehicles
• Key independent variable Average sectoral RoO
  restrictiveness of the 6-digit level final goods
  key products classified as final by section of
  HS
• Five sectors chemicals, machinery, TV and radio
  transmitters, textiles, vehicles

46
Regression Results Trade in Intermediates by
Sector and RoO
47
Preliminary Results Summary

• Both the restrictiveness and complexity of rules
  of origin reduce aggregate trade flows
• Regime-wide RoOand particularly the combined
  effects of such RoOdesigned to add leniency to
  the application of product-specific RoO boost
  aggregate trade
• Both the restrictiveness and complexity of rules
  of origin in FINAL goods significantly boost
  trade in INTERMEDIATES
• - Restrictive RoO in final goods may thus
  result in trade diversion in intermediate
  goods

48
Conclusions Future Clash or Harmonization of
RoO Regimes?

• EU and NAFTA-based RoO regimes will likely
  predominate in the future, particularly across
  Europe and the Americas
• Maybe an emerging Asian/trans-Pacific regime
  type?
• However, (1) harmonization of non-preferential
  RoO and
• (2) the possible de facto harmonization of
  preferential RoO due to the spread of the PANEURO
  and NAFTA models could facilitate future de jure
  harmonization of preferential RoO

49
Conclusions Why Harmonize Also Preferential
RoO?

• Countries operating in two or more RoO theaters
  simultaneously will
• Complicate firms supply relations
• Force specialization of production to a certain
  market(s) when RoO are strict
• Increase transaction costs when certification
  methods diverge
• This is particularly the case for small exporters
  in small countries that are spokes to different
  RoO hubs (PANEURO model, NAFTA model, etc.) even
  slight differences in RoO can have a big impact
  on production decisions
• Thus, Dohas mandating harmonization of
  preferential RoO could benefit smaller developing
  countries in particular

50
Conclusions How to Harmonize Preferential RoO
and Further Mitigate RoOs Impact ?

• Harmonized non-preferential RoO would provide a
  readily available baseline for harmonizing
  preferential RoO
• Preferential RoO are restrictive only as long as
  there are MFN tariffs the ultimate key to
  countering RoOs negative trade effects will lie
  in the success of multilateral liberalization